New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Introducing CNN-on-flash to ArmCL's layers #6

Closed

RedCarrottt opened this issue Nov 26, 2019 · 1 comment

Member

RedCarrottt commented Nov 26, 2019

CNN-on-flash seems to support only GEMM operators.

However, real machine learning apps usually use layer-wise operators; such as convolution layer, fully-connected layer.

If you want to apply it to convolution layer, supporting for im2col is imperative.

LeeHayun commented Dec 20, 2021

The current direction of research is mainly NAS that finds a deep learning model suitable for a given memory size rather than using less memory.

Therefore, it is more valuable to train a model for the GPU and generate an optimized GPU kernel than to apply the technique to the GPU.
These works are in progress in the following repositories.
https://github.com/SKKU-ESLAB/ANT-Model-DB
https://github.com/SKKU-ESLAB/Auto-Compression

So, I close this issue.

LeeHayun closed this as completed

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment